← Verified Truth Portal Русский

Living Proofs

Seven operational cases of the Verified Truth framework. Maturity calibrated. Iceberg-backed.

INTERNAL PREVIEW v0.1.1 · 2026-06-26 · audit-verified 36 supporting library docs

What this is

The Verified Truth framework is not one product. It is a substrate of twelve operational capabilities that compose into different verticals. The seven cases described below are the current Living Proofs of the framework — independent operational instances that demonstrate the substrate in different domains.

This document calibrates each case honestly: some are in Internal Demo, some in Pilot Discovery, some still at Concept. We use a seven-level maturity ladder rather than the binary "in production / not yet" framing that obscures more than it reveals.

Behind this presentation sits a research library of 44 documents (208 KB capability detail, 112 KB case detail, 160 KB industrial competitive landscape, 124 KB academic foundations). Every claim made here is intended to reference back to that substrate. The library is the iceberg. This presentation is the tip.

The maturity ladder

The framework uses a seven-level maturity model. Each case is positioned at its honest current level, with movement direction indicated where the case is in active transition (e.g. "Internal Demo → Pilot").

  1. 1
    Concept
  2. 2
    Research
  3. 3
    Prototype
  4. 4
    Internal Demo
  5. 5
    Pilot
  6. 6
    Production
  7. 7
    Commercial

A case at Concept has a deck or design but no implementation. A case at Internal Demo has running code accessible behind controlled access but no external partner adoption. A case at Pilot has a signed first user or partner. Production implies real ongoing operation at material scale. Commercial implies revenue-validated. None of the current cases have reached Production; the most mature is at the Internal Demo → Pilot transition.

01

EquiWork Agreements

Internal Demo → Pilot app.equiwork.io Cloudflare Access gated

The flagship deployment of the framework. Implements the full lifecycle from agreement creation through bilateral signing to canonical state for equity-compensation, service, consulting, and goods contracts. The architectural anchor that the other cases reference.

Capability coverage: 12 of 12 — the most capability-coverage-complete case. The architectural reference for the substrate.

Current state of evidence

  • Production-deployed lifecycle. The first canonical record was sealed on 2026-04-21. Five of the planned seven lifecycle states are sealed in production: Create → Freeze → Sign A → Sign B → Canonical with SHA-256 hash anchoring. (Executing and Completed are Roadmap.)
  • Codebase maturity. Approximately 80 API routes implemented. Auto-test suite passing 60 of 60 at last project record.
  • Wiring audit. 14 of 15 enterprise + multi-vertical features confirmed wired in the workspace component; the remaining one was fixed in a subsequent audit pass.
  • Operational stack. Production on Railway (backend) + Vercel (webapp) behind Cloudflare Access OTP gate. QA environment separately deployed.
  • Active P0 blockers. Split-brain architecture (CREATE writes to PostgreSQL, READ pulls in-memory seed data) and session persistence (sessions table exists but is unwired) remain pending CTO architectural decision. Confidential detail
Capability mapping (12 of 12)
01 IdentityInternal Demo → Production
02 ProvenanceProduction (lifecycle) / Internal Demo (audit portal UI)
03 EvidenceInternal Demo → Production
04 ReviewInternal Demo → Production
05 Human ApprovalProduction
06 OwnershipInternal Demo → Production
07 Version HistoryProduction
08 HandoffInternal Demo
09 AuditProduction (substrate) / Internal Demo (portal UI)
10 Access ControlInternal Demo → Production
11 AI Assistance (bounded)Internal Demo → Production
12 Export & InteroperabilityPrototype

Strengths

  • End-to-end lifecycle production-tested; the first canonical record is the proof-of-life event for the entire framework.
  • Architectural commitments enforced at the protocol level, not as policy — bounded AI, append-only audit, human approval all confirmed in production code.
  • The Telegram bot interface provides accessibility for non-technical signers without lowering the cryptographic integrity bar.

What this case does NOT claim

  • Not in production at scale. The architecture is sound at small scale; operational infrastructure has known gaps before scale.
  • Not a commercial deployment. No paying user as of this record.
  • Not regulatory-certified. Common Criteria, SOC 2, ISO 27001, GDPR compliance certifications are not held; the architectural commitments compose with them but no audit has been performed.
  • Not "patent-pending" per 35 USC §292 — proper Russian-language phrasing "в процессе патентного оформления" is used until a provisional is actually filed.
Cross-references: Research landscape · What is new here · Claims discipline
02

AI Hokusai

Internal Demo aihokusai.equiwork.io NDA + OTP gate

The first proof that the framework's substrate operates beyond agreement contracts. A cultural-provenance vertical for AI-assisted art creation, deployed as an ArtTech residency portal. The most explicit demonstration of the bounded-AI architectural commitment in any case.

Capability coverage: 11 of 12 — Ownership and Handoff are at Concept; the other ten capabilities are at Internal Demo or Prototype.

Current state of evidence

  • Seven interconnected views wired in the portal: Exhibition, Artist, Artwork Provenance, Residency CAP Ledger, Donor Transparency, Partner Accountability, Curator Portal.
  • NDA + OTP gate is production-deployed. A new visitor can request access, accept the NDA, receive an OTP via email, enter the portal. NDA acceptance is recorded as a real chain of attestations — the visitor's first verified record.
  • Live AI advisory layer proxying to Anthropic server-side. The API key never reaches the browser. AI generates only ANALYZED-class attestations; it structurally cannot generate APPROVE-class events.
  • Hanko (seal) aesthetic. Black-and-red elegant minimalism reflecting cultural-provenance subject matter rather than fintech aesthetics.
  • Pre-pilot. No first real residency partner or institutional adopter signed up; sample data populates the seven views.

Strengths

  • First proof that the framework is a platform, not a single product. Different vertical, same substrate, distinct visual identity.
  • Bounded AI is most explicitly demonstrated here. The portal uses AI advisory throughout the workflow while structurally preventing it from approving anything. AI advisory output is visibly tagged in the chain.
  • Custom domain and full deployment show that the framework's deployment pattern works for separate verticals (not just the monolithic main app).

What this case does NOT claim

  • Not authenticating AI-generated art. The portal records provenance — it does not adjudicate whether a work is "really" art or "really" by the credited human creator.
  • Not solving the AI-attribution problem. Many open legal and ethical questions remain (US Copyright Office, EU AI Act, ongoing litigation).
  • Not replacing gallery catalogues raisonnés or museum provenance research. It composes with them as the contemporary digital layer.
  • Not "the first AI-art provenance system" — adjacent systems exist (Verisart, Tagsmart, Arianee, Aura). The distinctness is the framework substrate, not the problem area.
Cross-references: Research landscape (industrial precedents: Verisart, Tagsmart, Arianee, Aura, Winston Artory+AXA)
03

Study Workspace

Internal Demo demo.equiwork.io Cloudflare Access OTP gated

A structured discernment and research environment that applies the substrate to scholarly and interpretive work. The case demonstrates that the framework operates not only for transactional records (agreements, object passports) but also for interpretive records — structured research outputs where the value lies in the integrity of the process rather than in a final certified truth.

Capability coverage: 11 of 12 (Ownership at Concept). Alongside EquiWork Agreements, the most capability-coverage-complete case.

Current state of evidence

  • Two operational wings. General (research cockpit, cool blue-graphite aesthetic) and Zohar (manuscript room, warm ivory parchment aesthetic, CSS-only Hebrew microtext decoration), sharing the same architectural substrate.
  • Demo Hub. 23-module system grid with explicit wing identities, Demo-vs-Real-Work comparison, "What a Reviewer Gets" section, bounded-AI advisory section.
  • Five-layer chip stack per event: Source / Commentary / Interpretation / Reflection / Review. Visible to the reviewer at the UI level.
  • Components in the codebase: ResearchDiagnostic, WorkingAnswerView, EvidenceTrail, ReviewCompletionRecordView, plus a pure computation library research.ts. Sixty-plus i18n keys in Russian and English. Implementation detail
  • Frontend-only prototype. Backend persistence would cross an architectural boundary that has been deliberately gated. Pre-pilot.

Strengths

  • The most explicit demonstration of "AI does not decide truth" in any case. The doctrine is operational, not just decorative.
  • The five-layer chip stack is a uniquely visible architectural pattern — the difference between Source, Commentary, Interpretation, Reflection, and Review is seen at the UI level.
  • Local-only design philosophy is honest about the framework's current operational boundary. The doctrine explicitly addresses the "hosted SaaS = crossing a boundary" framing.

What this case does NOT claim

  • Not certifying truth — philosophical, spiritual, factual, or legal. The doctrine explicitly rejects this.
  • The Zohar wing does NOT teach Kabbalah or Jewish doctrine. The wing is a research workspace for a specific interpretive subject; teaching is not the function.
  • Not an authority on interpretive questions. The workspace structures them for review; it does not resolve them.
  • Not yet adopted institutionally. No real research group has used the workspace for actual research.
Cross-references: Research landscape · What is new here (esp. the bounded-AI commitment)
04

Чёстно / Нечёстно (Honest / Dishonest)

Internal Demo game.equiwork.io Public access (no gate)

A serious game that trains recognition of verification patterns in real-world transactions. Pedagogically central to the framework: where the Study Workspace structures researchers' practice of structured verification, Чёстно/Нечёстно structures anyone's practice of recognising verification gaps in everyday consumer and professional contexts.

Capability coverage: 5–6 of 12 actively used. The remaining capabilities are taught within game scenarios rather than implemented operationally. This is the lowest capability-usage count of any case, by design — the value is pedagogical, not operational.

Current state of evidence

  • Seven languages live, including RTL support: Russian, English, Spanish, German, Polish, Ukrainian, Hebrew.
  • Three of thirteen planned roles have complete strategy cards in Russian and English: vintage broker (depersonalised from earlier draft), expert, fake-expert. Ten roles outstanding; five non-RU/EN language coverage gaps for existing roles.
  • Cache-busting discipline applied permanently — versioned query strings and Cache-Control headers, after a sequence of edge-cache incidents earlier in the project.
  • Database is ephemeral. SQLite that wipes on every Railway redeploy. No durable user progression. Persistent storage is the next operational gap. Operational detail

Strengths

  • Reach. Seven languages mean the case can speak to a much broader audience than any other vertical. Pedagogical reach is multiplicative.
  • Low barrier to entry. A game is more approachable than a contract platform or research workspace.
  • Demonstrates non-financial verticals. Most other cases have direct commercial value; this case demonstrates that the framework has educational and civic-value verticals.

What this case does NOT claim

  • Not certifying players in verification expertise. No certification offered or implied.
  • Not detecting or preventing fraud in the real world. The game trains recognition; players must exercise it in real contexts.
  • Not "first serious game in verification training" — multiple adjacent precedents exist (Foldit, Cyber Awareness Challenge, KnowBe4 training platforms).
  • The game does NOT itself use AI. It teaches about the bounded-AI commitment as a pattern to recognise.
Cross-references: Research landscape (serious games precedents: Foldit, Cyber Awareness, KnowBe4)
05

CAP Object Passport · Fur

Concept → Pilot Discovery Partner LOI received

A digital passport for vintage fur and vintage luxury items — not a certificate of authenticity, but a life record of an item: provenance, condition, expert inspection, storage, restoration, controlled access, sale, insurance, all in one place. The most thoroughly designed of the planned CAP verticals as of this record.

Capability coverage: 12 of 12 at Concept-to-Prototype maturity. The deck v0.1 demonstrates the most complete capability-coverage design of any case, despite the lowest implementation maturity. This is intentional — the deck is the partner-conversation document; implementation follows partner alignment.

Current state of evidence

  • Concept deck v0.1 — 10 pages, polished design, prepared for the partner conversation.
  • Letter of Intent received from the prospective vertical partner (subject-matter expertise, network, first batch of items for the pilot). Partner detail under NDA
  • Six-block record structure specified: basic info · photo protocol · condition · expert inspection · storage · restoration.
  • Eight-point photo protocol specified: collar and overall view; cuffs and sleeves; bottom and hem; lining and label; brand mark; seams and fastenings; problem areas / wear; fur close-up.
  • Six-level maturity ladder per item (the case's own ladder, distinct from the framework's case-level ladder): primary card → primary screening → basic passport → expert passport → passport for sale or insurance → collection passport.
  • Four-view access architecture: owner (full) · buyer (limited) · insurance (insurance package) · restorer (working access).
  • Pilot plan documented in deck: 5–10 real items of different types. Not yet executed.

Strengths

  • Calibrated honesty in the value proposition. "Not every old fur is valuable; here is honest screening" is far more credible than "we authenticate luxury goods". Framing reduces over-promise risk.
  • Explicit "what is not promised" section in the deck. The discipline the broader framework benefits from, demonstrated at the partner-conversation level.
  • Doctrine language already aligned. The deck uses the exact framework phrasing: система удостоверяет целостность процесса осмотра, а не «истину» вывода (the system certifies integrity of the inspection process, not the truth of the conclusion).
  • The most thoroughly designed access architecture of any case — the four-view model is a reference for other verticals.

What this case does NOT claim

  • Not a guarantee of authenticity. Authentication is performed by qualified experts; the framework records the inspection.
  • Not an appraisal or value oracle.
  • Not a legal guarantee or full protection from future claims.
  • Not an assertion that any old fur is valuable.
  • Not a publicly viewable database.
  • Not yet in pilot. Pilot Discovery is the honest framing — LOI exists, no items processed.
Cross-references: Research landscape (luxury digital passport precedents: Arianee, Aura, Tagsmart) · Claims discipline
06

CAP Object Passport · Wine

Pilot Discovery First conversation underway External-customer model

The wine-industry application of the CAP substrate. Operated through Tasting & Toasting Inc. (T&T), a separate Delaware Corporation that acts as the first external customer of the EquiWork CAP substrate rather than as an internal product. The case that demonstrates the framework's substrate-as-platform thesis at the entity level.

Capability coverage: approximately 12 of 12 at Concept level. The case is at Concept maturity for implementation; capability-usage estimates are projections from the substrate's design.

Current state of evidence

  • External-customer entity architecture operational. T&T is a separate Delaware Corporation; the CAP substrate is licensed; T&T deploys it for the wine vertical. The substrate-as-platform thesis is demonstrated at the entity level.
  • First winery conversation underway. Stara Winna Góra in Poland is the first prospective winery partner. The conversation is active; no signed agreement as of this record. Partner detail under NDA
  • Industry-aligned partner network through T&T's existing wine industry relationships.
  • No wine-specific deck or design document yet. Compare to the Fur case (deck v0.1 with full architectural specification). Building the wine-specific equivalent is the next deck-level work.

Strengths

  • External-customer entity architecture works. T&T as a separate Delaware Corporation customer of EquiWork CAP demonstrates the substrate-as-platform model at the entity level — not just architecturally but legally.
  • Industry-aligned partner. Prospective wineries see an industry peer, not a generic tech vendor.
  • High per-unit value at the premium end. Collector wines can be five-figure or six-figure per bottle, making per-item provenance economically viable.
  • Cross-pollination with the Fur case. The deck v0.1 design patterns can be adapted to wine without redesign.

What this case does NOT claim

  • Not in pilot. Pilot Discovery is honest framing; no signed agreement.
  • Not authenticating vintage wine. Authentication is performed by qualified experts; the framework records the authentication act.
  • Not detecting counterfeit wine. Counterfeit detection is performed by chemical analysis, label forensics, expert visual inspection.
  • Not handling alcohol regulatory compliance. Per-jurisdiction compliance is per-deployment work.
Cross-references: Research landscape (wine precedents: eProvenance, Selinko, Authentic Vision, Aura wine partners)
07

CAP Object Passport · Paintings

Concept Planned via Cartulary Inc. Insurer-led GTM

The strategic vertical wedge for fine-art conservation, provenance, and insurance — planned for deployment through a separate Delaware C-Corp (Cartulary Inc.) that would license the CAP substrate from EquiWork. The most commercially-planned vertical despite having the lowest implementation maturity. The framing is honest about that asymmetry.

Capability coverage: approximately 12 of 12 at Concept level. The architectural design is comprehensive; the implementation is zero. The case is plan-rich, code-poor.

Current state of evidence

  • Six-tier architecture designed: T1 (private owner) through T6 (insurer white-label). Allows the substrate to operate at multiple scales from individual collector to multi-billion-dollar insurer book of business.
  • Go-to-market identified. Insurer-led adoption with target insurer set named: Chubb, Hiscox, AIG Private Client, Lloyd's syndicates.
  • Competitive event documented. AXA was eliminated from the target insurer list following the AXA XL / Winston Artory Group strategic partnership announced August 14, 2025. (Per public reporting: Casey Santangelo, Head of Fine Art & Specie Insurance Americas at AXA XL; Elizabeth von Habsburg, Co-Founder & Co-Executive Chair at WAG.)
  • Cloud infrastructure decision: GCP from day one, reflecting the operational and compliance demands of regulated insurance industry engagement.
  • Entity formation status uncertain for Cartulary Inc. The plan is documented; current entity state needs confirmation. Strategic detail
  • Zero implementation. No code written. No insurer conversations begun. No conservator partners. No gallery partners.

Strengths

  • The market is well-defined. Fine art insurance is a concentrated, mature, high-margin market. The pain points (provenance verification, claim evidence, conservation history) are documented and well-understood.
  • Insurer-led GTM is the right architecture. Insurers are the buyer with the operational pain; pulling through to galleries, conservators, and owners follows.
  • Competitive landscape is mapped. The case knows what Verisart, Tagsmart, Arianee, Aura, and Winston Artory do; the differentiation is articulated.
  • The strategic framing matches the framework's posture. The case does not promise art authentication; it promises substrate for the acts that establish authentication and conservation.

What this case does NOT claim

  • Not in pilot. No conversations have begun. Concept is honest.
  • Not authenticating paintings. Authentication is performed by qualified experts.
  • Not replacing conservator expertise. Conservators perform conservation; the framework records the act.
  • Not replacing insurer underwriting. Underwriters perform underwriting; the framework records supporting evidence.
  • Not "patent-pending" per 35 USC §292.
  • Not operational. There is no Cartulary Inc. product deployed. Even the entity formation status is uncertain.
  • Not "the leading platform for fine art provenance" — competitive landscape is well-populated; the leadership claim is unsupportable.
Cross-references: Research landscape (art provenance: Verisart, Tagsmart, Arianee, Aura, Winston Artory+AXA)

The pattern across the seven cases

Seven cases with seven different maturity levels are not a weakness of the framework. They are evidence of the substrate-as-platform thesis: a single set of twelve operational capabilities composing into seven distinct verticals, each at the level of maturity appropriate to its current investment.

The substrate is reusable. All seven cases share the same twelve capabilities. The capability docs (208 KB of detailed specification) are written once; each case references back. Verticals don't reinvent identity, audit, access control, bounded AI — they compose with the substrate.
The bounded-AI commitment is architectural, not policy. Three cases (EquiWork Agreements, AI Hokusai, Study Workspace) operationally demonstrate the commitment. In all three, AI can assist but cannot transition records to canonical state. The commitment is protocol-level enforced, not "best-practice" suggested.
The maturity-honest framing is institutional, not stylistic. Every case has explicit "What this does NOT claim" boundaries. This is not modesty rhetoric — it is the framework's claims-discipline operational at the case level. The same discipline applies in capability docs, in industrial-precedent docs, in the four research documents in the /research/ subdirectory.
External-customer entities are part of the architecture. T&T (wine vertical) and the planned Cartulary Inc. (paintings vertical) are separate Delaware corporations that license the CAP substrate. This is not white-labelling — it is substrate-as-platform at the entity level. The substrate scales by external-customer adoption, not just by internal product expansion.
The framework composes with industry rather than competing. Where existing systems work (DocuSign for signing, Cloudflare for access gating, ORCID for academic identity, IBM Food Trust for supply chains, Sigstore for software signing), the framework composes with them. Differentiation is at the lifecycle-substrate level, not at the per-component level.
The iceberg matters. This presentation is 50 KB. The library behind it is 604 KB across 44 documents — 12 capability specifications, 7 case documents, 17 industrial-precedent docs, 8 academic-foundation docs. Every claim in this presentation references back. The depth of the library is what permits the calibrated brevity of the presentation.

What this presentation is NOT

  • Not a pitch deck. This material exists for partners, advisors, and the team to understand the framework's operational reality. It is not optimised for investor-attention or sales-velocity.
  • Not a marketing site. Every claim is calibrated against the underlying research library. Marketing rhetoric ("the world's first", "the leading", "the only") is structurally absent because the discipline forbids it.
  • Not a product catalogue. The seven cases are not seven products for sale. They are operational instances of the framework at different maturity levels.
  • Not exhaustive. Future verticals may emerge (the Scientific Knowledge Passport concept; potential industry verticals around aviation parts provenance, biotech reagent provenance, professional credentialing). They are not in this document because they are not yet at Concept-deck level.
  • Not unchanging. Each case's maturity will evolve. This is v0.1; we will issue v0.2 when material changes warrant it (a first paying user moves a case from Internal Demo to Pilot; an insurer engagement moves Paintings from Concept; etc.).

Further reading

The four research documents in the /research/ subdirectory are the calibrated entry points to the broader research library:

  • Research landscape — the framework's position in eight intellectual fields with explicit "where EquiWork extends or differs"
  • What is new here — seven structural moves the framework makes
  • Claims discipline — sixty-three enumerated claims classified, with explicit "Forbidden Claims" section
  • References — forty-five web-verified bibliographic entries

Behind these public documents sits the internal library (44 docs, 604 KB) including the per-capability operational specifications, the per-precedent competitive analysis, and the per-foundation academic grounding. Access to the library is available to scientific reviewers, strategic advisors, and partners under appropriate access tier.